There is one question underneath all of this and it is the hardest to ask without sounding either glib or bitter, so I am going to set it down now and leave it until the end.
On Wednesday the order was given for the remaining American forces in Syria, something on the order of a thousand people, to leave within two months. They are advisers, trainers and the support staff who keep advisers and trainers supplied. The withdrawal is not finished; it is instructed. A week earlier the garrison at al-Tanf, in the south east near the Iraqi and Jordanian borders, was handed to Syrian government control.
The ground presence began in 2014, under the coalition campaign against Islamic State. Twelve years, ending on a schedule measured in weeks.
A thousand people is a small number and it is worth being clear about what a small number of that kind does, because the headline makes it sound like a garrison. Advisers sit with a partner force’s commanders and plan operations with them. Trainers run the courses that turn recruits into units. The support tail keeps both supplied, medevaced and connected to the intelligence and air power that is the reason a partner accepts advisers in the first place. None of that is combat power in itself. All of it is the connective tissue that lets a local force operate at a level it could not otherwise sustain, which is precisely why removing it changes what that force can do.
What follows is not an argument about whether to go. It is an accounting: what the twelve years cost, who paid it, and who is holding the balance now that the party that ran the operation is leaving.
What it cost, and to whom
The American cost is the one that gets counted, and by the standards of the wars that preceded it, it was modest. A deployment that peaked in the low thousands and latterly sat near a thousand. No occupation, no nation building budget, no draft. That modesty was the entire design: a small footprint, working through a local partner, at a price a democracy would tolerate indefinitely.
The partner is where the cost actually landed.
The Kurdish led Syrian Democratic Forces provided the infantry for the campaign against Islamic State. They took the territory and they took the casualties, in the thousands, over years, in the fighting that broke the group’s control of its cities. That was the arrangement and both sides understood it: American air power, intelligence and equipment, and somebody else’s people walking into the buildings.
It is not a criticism of either party to say so. It is the shape of the thing, and the shape determines who is exposed when it ends.
What the partner is holding now
By January the SDF had lost substantial ground to Syrian government forces in fighting across the north and east, and has since come to an arrangement to integrate into the army of that same government. A force built and equipped as a partner of the United States against Islamic State is being folded into the Syrian state, on terms set at a moment when its principal backer had announced it was going.
That is the transfer at the center of this ledger, and it is the kind that does not appear in any budget.
The detainees
The most concrete liability is a population of people in prisons.
The SDF has held thousands of Islamic State fighters, including foreign nationals whose home countries have spent years declining to take them back. Those prisons are not purpose built facilities with settled legal status. They are improvised, chronically underfunded, and guarded by a force that has just been through a war with its own government.
The fragility is not theoretical. In January the SDF lost control of the prison at al-Shaddadah after an armed assault and a number of detainees escaped. Earlier this month, American vehicles escorted buses carrying detainees out of the north east into Iraq, which is a partial answer to the question of who takes custody and is not a complete one.
The uncounted cost here is a specific kind: a problem that was manageable because somebody was paying attention and funding it, transferred to actors with less money, less capacity and more immediate concerns. It does not produce a bill. It produces a probability.
Who collects
The Syrian government collects territory, including al-Tanf, and collects the integration of a force that was until recently a rival authority over a third of the country. Turkey collects the departure of an American security guarantee over an armed Kurdish formation on its border, which has been its stated objective throughout. Russia and Iran collect a reduced American presence in a country where both have interests and neither has left.
Islamic State collects the opportunity presented by weakened prisons and a partner force preoccupied with its own absorption. That is not a prediction that the group reconstitutes. It is a statement that the conditions that made reconstitution hard have got worse.
The comparison worth making
This is not happening alone. The American withdrawal from Iraq has been running through the same period, on its own timetable, for its own reasons, and the two together describe a single decision about the region rather than two local ones.
The Iraqi case is the more instructive because it has been done before. American forces left Iraq at the end of 2011 and returned in 2014, and the thing that brought them back was precisely the collapse of a local partner against a group that had been contained rather than destroyed. Nobody planned that return. It was produced by the gap between a force being defeated and a force being finished.
Whether that pattern repeats is genuinely open, and anyone confident either way is guessing. The useful point is narrower: a withdrawal is a bet that the local arrangement holds, and the last time this bet was placed in this region it was lost inside three years, at a cost far exceeding what staying would have run to.
The question
So: what was it for?
The honest answer is that it achieved its stated objective and never acquired a second one. Islamic State’s territorial control was destroyed, and that was real, and the people who did it were mostly not American. What followed was twelve years of holding a position with no defined end state, because nobody could say what condition would have to be true for the job to be finished, and in the absence of such a condition the presence simply continued until a government decided it had had enough.
That is not a scandal. It is the ordinary fate of open ended commitments, and it is why the departure looks abrupt: there was never a marker to arrive at, so any date would have looked arbitrary, and this one is.
The people who will find out whether the twelve years were worth it are not the ones who spent them.
